action #34282
closed
action #34042: [tools] Worker goes to infinite loop during upload of screenshots in case of writing failure
implement catching errors in all our log_* functions
Added by szarate about 6 years ago.
Updated over 5 years ago.
Category:
Regressions/Crashes
Description
Currently the logging functions that we have in our codebase, do not handle errors in most cases, as they are used everywhere in the code and would prevent cleanup phases from being executed if they fail to print a message.
A fallback to croak or cluck if all else fails would be a good starting point.
- Related to action #40862: Out of disk space killed the webui (on osd) added
- Category set to Regressions/Crashes
- Priority changed from Normal to High
- Target version set to Ready
This sounds important enough to handle - we can't be killed by logging
- Related to deleted (action #40862: Out of disk space killed the webui (on osd))
- Has duplicate action #40862: Out of disk space killed the webui (on osd) added
- Target version changed from Ready to Current Sprint
picking it for current sprint
- Status changed from New to In Progress
- Status changed from In Progress to Feedback
PR is merged. Let's see how well it works in production.
- Status changed from Feedback to Resolved
Seems like it didn't break anything. If logging fails on production again and my changes turn out to be insufficient we can reopen the ticket.
- Target version changed from Current Sprint to Done
Also available in: Atom
PDF